• Corpus ID: 207853138

Learning to reinforcement learn for Neural Architecture Search

  title={Learning to reinforcement learn for Neural Architecture Search},
  author={J. Gomez Robles and Joaquin Vanschoren},
Reinforcement learning (RL) is a goal-oriented learning solution that has proven to be successful for Neural Architecture Search (NAS) on the CIFAR and ImageNet datasets. However, a limitation of this approach is its high computational cost, making it unfeasible to replay it on other datasets. Through meta-learning, we could bring this cost down by adapting previously learned policies instead of learning them from scratch. In this work, we propose a deep meta-RL algorithm that learns an… 

Deep Reinforcement Learning Versus Evolution Strategies: A Comparative Survey

An overview of how DRL and ESs can be used, either independently or in unison, to solve specific learning tasks is presented and is intended to guide researchers to select which method suits them best and provides a bird's eye view of the overall literature in the field.

Weight-Sharing Neural Architecture Search: A Battle to Shrink the Optimization Gap

A literature review on the application of NAS to computer vision problems is provided and existing approaches are summarized into several categories according to their efforts in bridging the gap.

Accuracy-Time Efficient Hyperparameter Optimization Using Actor-Critic-based Reinforcement Learning and Early Stopping in OpenAI Gym Environment

Accuracy-time efficient hyperparameter optimization (HPO) using advantage actor-critic (A2C)-based reinforcement learning (RL) and early stopping in OpenAI Gym environment and results show that the presented accuracy-efficient HPO architecture can improve 0.77% accuracy on average compared with defaulthyperparameter for random forest.

Pruning Filter via Gaussian Distribution Feature for Deep Neural Networks Acceleration

A novel compression and acceleration method based on data distribution characteristics for deep neural networks, namely Pruning Filter via Gaussian Distribution Feature (PFGDF), which compresses the model by filters with insignificance in distribution, regardless of the contribution and sensitivity information of the convolution filter.

Active learning using adaptable task-based prioritisation

A controller neural network that measures priority of images in a sequence of batches, as in batch-mode active learning, for multi-class segmentation tasks, and shows that the proposed adaptable prioriti- sation metric yields converging segmentation accuracy for the novel class of kidney, unseen in training.



Learning to reinforcement learn

This work introduces a novel approach to deep meta-reinforcement learning, which is a system that is trained using one RL algorithm, but whose recurrent dynamics implement a second, quite separate RL procedure.

Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks

We propose an algorithm for meta-learning that is model-agnostic, in the sense that it is compatible with any model trained with gradient descent and applicable to a variety of different learning

A Study of the Learning Progress in Neural Architecture Search Techniques

The results indicate that one-shot architecture design is an efficient alternative to architecture search by ENAS, and the learning curves are completely flat, i.e., there is no observable progress of the controller in terms of the performance of its generated architectures.

Neural Architecture Search with Reinforcement Learning

This paper uses a recurrent network to generate the model descriptions of neural networks and trains this RNN with reinforcement learning to maximize the expected accuracy of the generated architectures on a validation set.

RL$^2$: Fast Reinforcement Learning via Slow Reinforcement Learning

This paper proposes to represent a "fast" reinforcement learning algorithm as a recurrent neural network (RNN) and learn it from data, encoded in the weights of the RNN, which are learned slowly through a general-purpose ("slow") RL algorithm.

Designing Neural Network Architectures using Reinforcement Learning

MetaQNN is introduced, a meta-modeling algorithm based on reinforcement learning to automatically generate high-performing CNN architectures for a given learning task that beat existing networks designed with the same layer types and are competitive against the state-of-the-art methods that use more complex layer types.

Playing Atari with Deep Reinforcement Learning

This work presents the first deep learning model to successfully learn control policies directly from high-dimensional sensory input using reinforcement learning, which outperforms all previous approaches on six of the games and surpasses a human expert on three of them.

Efficient Neural Architecture Search via Parameter Sharing

Efficient Neural Architecture Search is a fast and inexpensive approach for automatic model design that establishes a new state-of-the-art among all methods without post-training processing and delivers strong empirical performances using much fewer GPU-hours.

Regularized Evolution for Image Classifier Architecture Search

This work evolves an image classifier---AmoebaNet-A---that surpasses hand-designs for the first time and gives evidence that evolution can obtain results faster with the same hardware, especially at the earlier stages of the search.

Reinforcement learning in robotics: A survey

This article attempts to strengthen the links between the two research communities by providing a survey of work in reinforcement learning for behavior generation in robots by highlighting both key challenges in robot reinforcement learning as well as notable successes.